Quantitative Analysis of Iris Changes Following Mydriasis in Subjects With Different Mechanisms of Angle Closure

Abstract

Purpose.: We estimate and compare change in iris cross-sectional area (IA) after physiologic and pharmacologic mydriasis in subjects with different dominant mechanisms for primary angle closure.

Methods.: Anterior segment optical coherence tomography (AS-OCT) measurements in light, dark, and following pharmacologic dilation were obtained on primary angle closure suspects (PACS), primary angle closure (PAC), and primary angle closure glaucoma (PACG) examined during the 5-year follow-up of the Handan Eye Study. Subjects were categorized into three subgroups according to their dominant angle closure (AC) mechanisms as determined by AS-OCT: pupillary block (PB), plateau iris configuration (PIC), and thick peripheral iris roll (TPIR). The IA and other biometric parameters measured using the Zhongshan Angle Assessment Program in the right eyes of all subjects were analyzed.

Results.: A total of 364 right eyes of subjects with PACS (333), and PAC/PACG (31) was included in the analysis. Significant differences in the change of IAs (P = 0.030), IA loss per mm pupil diameter (PD) increase (P = 0.001) in light versus pharmacologic dilation, and IA loss per mm PD increase (P = 0.011) from dark versus pharmacologic dilation were observed among the three groups. The smallest decrease occurred in the PB group.

Conclusions.: There are significant differences in IA and IA loss per mm of pupil change following physiologic or pharmacologic mydriasis in Chinese subjects with dissimilar dominant mechanisms for AC. Dynamic iris change may have a more important role in angle closure where PB is the dominant mechanism.

Introduction

The International Society of Geographical and Epidemiological Ophthalmology (ISGEO) classification for primary angle closure (PAC) includes primary angle closure suspect (PACS), PAC, and primary angle closure glaucoma (PACG).^1,2 The PACS essentially is a risk factor for disease and is defined as eyes where the posterior trabecular meshwork is not visible for at least 180°, IOP ≤ 21 mm, no peripheral anterior synechiae (PAS) with healthy optic nerves.³ The presence of raised IOP (>21 mmHg) and/or PAS in a PACS is termed PAC, while the term PACG is used only in the presence of glaucomatous optic neuropathy (GON)/visual field loss with PAC.^1,2 In this report, we refer to PAC and PACG as primary angle closure disease (PACD). 

Recent reports support the concept that PAC is a multifactorial disease caused by a combination of anatomical and dynamic (iris and choroid) components.^4–7 While anatomical factors have been extensively investigated, the role of dynamic elements is less clear.^8–12 Analysis of such dynamic iris behavior may provide clues to understanding which eyes might develop angle closure (AC) and perhaps explain the higher prevalence of PACD among Asians, as compared to Europeans and Africans.^8,13–16 

Using anterior segment optical coherence tomography (AS-OCT), Quigley et al.⁸ reported that a smaller change in iris cross-sectional area (IA) with pupil dilation could be a potential risk factor for AC in those with European ancestry. They hypothesized that a smaller decrease in iris volume (IV) on dilation may present a higher risk of AC.⁸ In South Indians, IA and IV decreased with pupillary dilatation in normal and AC eyes; the loss of IV was lower with AC.¹⁰ It is hypothesized that high fluid content of the iris stroma combined with a reduction in the transfer of extracellular fluid to the anterior chamber due to compact or water-retentive stroma could affect change in IA and IV as the pupil dilates, a dynamic factor that could predispose some anatomically susceptible eyes to PAC.^{8–10,17–20} Our previous research in a rural Chinese population showed that PACS and normal eyes respond differently to physiologic and pharmacologic pupillary dilation, with the former showing a smaller reduction in IA and IV.²¹ Such differences in iris behavior may partly explain why only a small proportion of PACS eyes develop PACD. 

Aptel and Denis⁹ reported that IV estimated with the AS-OCT, increased after pupil dilation in narrow-angle eyes predisposed to acute angle closure (AAC). They also found that this biometric change was associated with angle narrowing, despite a patent laser peripheral iridotomy (LPI).⁹ However, the increase in iris volume in some eyes is likely to be due to a fallacy in the formula used for its calculation.²² Narayanaswamy et al.²⁰ reported that the IV decreased in eyes with chronic AC following physiologic mydriasis, but remained unchanged in fellow eyes of AAC. These findings suggested that variations in iris responses may influence the subtype of AC that develops and perhaps have more of a role in AAC. A recent study reports that eyes with loss in IA, but a paradoxical increase in IV had larger increase in centroid-to-centroid distance (the distance between the centers of right and left iris masses, CCD) relative to the increase in pupil size, than did the eyes with IA and IV loss.²² Hence, IA may be a better parameter that represents iris dynamic change, compared to IV change. 

Chinese eyes too seem to have multiple mechanisms for AC.^23–25 Using ultrasound biomicroscopy (UBM) Wang et al.²³ reported the following mechanisms for AC in Chinese eyes: pure pupillary block (38.1%), pure nonpupillary block (7.1%), and multiple mechanisms (54.8%); multiple mechanisms were the commonest. The AC mechanism involved in most AAC is predominantly pupillary block (PB), while nonpupillary block or multiple mechanism have a greater role in nonacute presentations.^26,27 In complex disorders like PAC, all individuals with disease do not share the same set of risk factors and the framework of a sufficient component causal model is useful in identifying individual mechanisms of causation.⁶ While pupillary block is a possible universal necessary component of all sufficient component cause models of PAC, identification of other component mechanisms that can be addressed is likely to be clinically useful.⁶ We hypothesized that the contribution of dynamic iris behavior to the pathogenesis of PACD patients may vary among those with different AC mechanisms. 

The objective of this study was 2-fold: (1) Quantify changes in IA induced by physiologic and pharmacologic mydriasis in subgroups with different mechanisms of AC as determined by AS-OCT and detect any differences between them. (2) Study the association of such changes with demographic factors, previously reported ocular biometric measurements as well as the different mechanisms responsible for AC. 

Methods

Subjects and Ophthalmic Examination

Patients for this observational, cross-sectional study were selected from the 5-year follow-up of the Handan Eye Study (HES). The HES was conducted on a sample of rural Chinese adults aged 30 years or older living in Handan County, Hebei Province.²⁸ From May 2012 to June 2013, surviving members of the original HES cohort were re-examined for the 5-year follow-up.²¹ 

The HES subjects aged ≥40 years who participated in this follow-up examination between September 2012 and May 2013, underwent gonioscopy and had an occludable angle (the posterior trabecular meshwork not visible for at least 180°) were eligible for inclusion. Gonioscopy was performed on all subjects with a peripheral limbal anterior chamber depth less than or equal to 40% of peripheral corneal thickness as well as for 1 in 10 subjects (number 1, 11, 21, and so forth) registered per day.²¹ Subjects with previous intraocular surgery, previous penetrating eye injury, or corneal disorders preventing anterior chamber assessment and persons taking antiglaucoma eye drops were excluded, as were those who had suffered an episode of AAC or had undergone LPI or laser iridoplasty. Subjects who were on topical or systemic medication that could affect the iris or angle configuration at the time of the study (cholinergics or anticholinergics, adrenergic agonists or antagonists, serotonin, norepinephrine, and dopamine releasers, their precursors or reuptake inhibitors, monoamine oxidase inhibitors; opioid agonists or antagonists, and histamine receptor antagonists) also were excluded.²¹ 

All participants underwent a comprehensive ophthalmic examination including presenting (PVA) and best corrected visual acuity (BCVA), using the Early Treatment Diabetic Retinopathy Study (EDTRS) LogMAR E chart, objective and subjective refraction, slit-lamp biomicroscopy, visual field examination, applanation tonometry, gonioscopy, A-scan ultrasound biometry, and fundus examination. Refraction was measured using a KR-8800 auto kerato-refractometer (Topcon, Tokyo, Japan), visual filed tested using the 24-2 Swedish Interactive Testing Algorithm (SITA) standard program on a visual filed analyzer (Humphrey Visual Field Analyzer 740i or 750i; Carl Zeiss, Jena, Germany), and A-scan ultrasound biometry using an OcuScan RxP (Alcon, Inc., Fort Worth, TX, USA).²¹ Static gonioscopy was performed at high magnification (×25) with the eye in the primary gaze position using a Goldmann-type one-mirror lens under the lowest level of ambient illumination that permitted a view of the angle.²¹ Dynamic examination (manipulation) then was performed using the same lens. Gonioscopy was performed by one of two observers who were masked to AS-OCT findings.²¹ The two observers attained a Kappa (κ) of 0.76 for assessment of occludable angle in 30 eyes (not included in this study).²¹ 

The study was approved by the ethics committee of the Beijing Tongren Hospital and performed in accordance with the tenets of the Declaration of Helsinki. All subjects provided verbal and written informed consent. 

AS-OCT Imaging

The AS-OCT study uses an infrared light with a wavelength of 1310 nm that optimizes anterior chamber angle imaging in the absence of visible light spectrum influence on angle configuration and pupil size.^29,30 This technique enables cross-sectional images of the anterior segment of the eye and is capable of recording transient and dynamic changes of the pupil at low levels of illumination.^29,30 Each eye was imaged with an AS-OCT (Visante; Carl Zeiss Meditec, Inc., Dublin, CA, USA), first under dark conditions (approximately 3 lux, to induce physiologic mydriasis), then with light (approximately 200 lux), and finally 30 minutes after pharmacologic dilation with tropicamide 1% eyedrops. Pharmacologic pupil dilation was not performed on subjects with a broad PAS (≥ 6 clock hours) because of the high risk of AAC. 

Since PACS and PAC/PACG eyes have a risk of AAC post dilation, the protocol incorporated precautions to recognize and manage such events. Subjects recording a normal IOP (<21 mm Hg) at least 1 hour after dilation were allowed to leave.²¹ Those with an IOP ≥ 21 mm Hg received IOP-lowering medications as required.²¹ The protocol required those with raised IOP at high risk for an acute event to stay in the central clinic for one night for further observation and management.²¹ Doctors in the surrounding towns and villages were made aware of the symptoms that required patients to be sent back to the clinic.²¹ 

All images were obtained in the “anterior segment quadrant” mode at 0° to 180°, 45° to 225°, 90° to 270°, and 135° to 315° meridians. For image acquisition at 6 and 12 o'clock, the operator gently retracted the upper and lower lids as needed taking care to avoid inadvertent pressure on the globe.²¹ Imaging was repeated if the scleral spur visibility was poor, and the best set of images were selected.²¹ To obtain images in a nonaccommodated state, the subject's refractive correction was used to adjust the internal fixation target for their distance correction.²¹ The scleral spur was marked by one of us (ZY) and custom software (Zhongshan Angle Assessment Program; ZAAP, Guangzhou, China) was used to measure IA and other parameters.³¹ This investigator was trained to identify the scleral spur and has performed measurements using ZAAP on approximately 10,000 AS-OCT images. 

Angle and anterior chamber configuration, including angle opening distance at 500 μm (AOD500), trabecular-iris space at 500 μm (TISA500), angle recess area (ARA), anterior chamber depth (ACD), anterior chamber area (ACA), anterior chamber volume (ACV), anterior chamber width (ACW), and pupil diameter (PD) also were analyzed with the same software. An AOD at 500 μm is the distance from the corneal endothelium to the iris surface as determined from a perpendicular to a line drawn at 500 μm from the scleral spur.³² The TISA500 is the area bounded anteriorly by the AOD500 as determined, posteriorly by a line drawn from the scleral spur perpendicular to the plane of the inner scleral wall to the iris, superiorly by the inner corneoscleral wall, and inferiorly by the iris surface.³³ The ARA is the area bordered by the anterior iris surface, corneal endothelium, and a line perpendicular to the corneal endothelium drawn to the iris surface from a point at 750 μm anterior to scleral spur.³⁴ 

Categories of Angle Closure Mechanisms

Four AS-OCT images from each eye obtained in the dark with clearly discernible scleral spurs were analyzed qualitatively and categorized into one of three AC mechanisms: PB, plateau iris configuration (PIC), and thick peripheral iris roll (TPIR). Where the image suggested more than one mechanism for AC, one of us (ZY) made a forced choice to select the dominant mechanism without the benefit of other information. The AC mechanism that was identified in at least two AS-OCT images of each eye was determined to be predominant AC mechanism of that eye. In 40 PAC eyes (at the Beijing Tongren Hospital) that underwent UBM and AS-OCT, the κ for such a forced choice of AC mechanism between the instruments was 0.87 (unpublished data). 

The guidelines to categorize the images into three AC mechanisms were as follows: (1) PB (Fig. 1): Convex forward iris profile, giving the typical bombe appearance, a very small zone of iris-lens contact in the center and shallow peripheral anterior chamber. (2) PIC (Fig. 2): The peripheral iris rises up from its root in apposition or in close proximity to the angle wall and then turns sharply away from the angle toward the visual axis and central anterior chamber is relatively deep, but the periphery is shallow. (3) TPIR (Fig. 3): Eyes with this condition have a thick iris, which is thrown into prominent peripheral circumferential folds occupying a large proportion of the angle. The central anterior chamber is relatively deep but the periphery is shallow.^35–38 

Figure 1

 

Angle closure mechanism of pupillary block.

Figure 2

 

Angle closure mechanism of plateau iris configuration.

Figure 3

 

Angle closure mechanism of thick peripheral iris roll.

Statistical Analysis

The IA and PD values represented the average of measurements from eight iris cross-sections as obtained from AS-OCT scans; the terms IA and PD are used to designate these average values. Only the right eye of each subject was included for analysis. The 1-sample Kolmogorov–Smirnov test was used to assess normality of the measurements. Variables demonstrating a normal distribution are presented as mean (SD), while variables failing to achieve a normal distribution are presented as median (percentiles). The ANOVA, nonparametric tests, and χ²-test were used to compare differences among the three groups. Linear regression analysis was used to analyze factors associated with change in IA. Univariable regression was conducted with changes in IA as the dependent variable, and the effects of age, sex, central corneal thickness (CCT), ACD, lens thickness (LT), axial length (AL), PD in light, PD in dark, PD change, CCD in light, CCD in dark, CCD change, and AC mechanisms as predictors. Variables that were significant at a level of P < 0.05 were included in a multivariable linear regression model. The SPSS statistical software version 17.0 (SPSS, Inc., Chicago, IL, USA) was used for data analysis. Statistical significance was set at P < 0.05. 

Results

Subjects Characteristics

A total of 520 subjects (476 PACS and 44 PAC/PACG) attending the 5-year HES follow-up were eligible for inclusion; 79 eyes (15.2%) were excluded due to poor image quality and inability to accurately identify the scleral spur. There were no statistically significant differences in demographic or ocular features between included eyes/patients and excluded eyes/patients. We excluded 10 eyes considered to have an exaggerated lens vault (Fig. 4) as the AC mechanism and 15 eyes in which the dominant AC mechanism could not be identified. As a small change in pupil size would not contribute to, or even mask information, 52 eyes with pupil dilation less than 0.5 mm from light to dark also were excluded. Following these exclusions, 364 eyes (333 PACS and 31 PAC/PACG) were available for analysis; 110 eyes were determined to have PB (29.3%), 125 eyes PIC (36.1%), and 129 eyes TPIR (34.6%) as the dominant mechanism. Considering a PD change as ≥0.5 mm, data from 278 eyes were available for change from light to dark and for 261 and 264 eyes from change from light or dark to pharmacological pupil dilation. 

Figure 4

 

Angle closure mechanism of exaggerated lens vault.

The demographic and ocular biometric data of subjects with each AC mechanism group are shown in Table 1. There was no significant difference in age, sex, BCVA, spherical equivalent (SE), IOP, and CCT among the three groups. A significant difference existed in PVA (P = 0.019), ACD (P = 0.001), LT (P = 0.039), and AL (P = 0.041) among the three AC mechanisms. Eyes with PIC had better PVA than eyes with PB (P = 0.012), and eyes with PIC had deeper ACD than eyes with PB and TPIR (P = 0.002 and 0.013) as well as a smaller LT (P = 0.018) compared to the PB group. 

Table 1

 

Demographic Data and Ocular Biometric Measurements of Subjects With Different PAC Mechanisms

Parameter	1 = PB,n= 110	2 = PIC,= 125	3 = TPIR,n= 129	P Value	P Value, 1 & 2	PValue, 1 & 3	PValue, 2 & 3
Age, y (SD)	63.4 (8.4)	60.9 (7.7)	62.0 (8.7)	0.062*
Male (%)	31 (28.2)	46 (36.8)	42 (32.6)	0.372†
Female (%)	79 (71.8)	79 (63.2)	87 (67.4)
PVA (IR)	0.32 (0.20, 0.40)	0.20 (0.10, 0.40)	0.30 (0.14, 0.44)	0.019‡	0.012‖	0.425‖	0.026‖
BCVA (IR)	0.10 (0.00, 0.20)	0.00 (0.00, 0.12)	0.07 (0.00, 0.17)	0.137‡
SE (SD)	0.92 (1.48)	0.79 (1.34)	0.82 (1.31)	0.810*
IOP (IR)	12.0 (10.0, 13.5)	12.0 (10.0, 13.5)	12.0 (10.0, 13.0)	0.672‡
CCT (SD)	532 (28)	537 (29)	531 (31)	0.205*
Central ACD (SD)	2.51 (0.24)	2.63 (0.29)	2.53 (0.29)	0.001*	0.002§	1.000§	0.013§
LT (IR)	4.90 (4.56, 5.12)	4.75 (4.31, 5.02)	4.89 (4.44, 5.12)	0.039‡	0.018‖	0.699‖	0.048‖
AL (IR)	22.21 (21.76, 22.92)	22.39 (21.91, 23.18)	22.28 (21.69, 22.28)	0.041‡	0.034‖	0.983‖	0.026‖

 IR, interquartile range.

*  A 1-way ANOVA.

†  The χ² test.

‡  Kruskal-Wallis test.

§  Bonferroni test.

‖  Mann-Whitney test (<0.05/3 = 0.017 = significantly different).

Quantitative anterior chamber parameters measured using AS-OCT in the three AC mechanisms in light, dark, and after pharmacologic dilation, along with the differences between the groups are summarized in Tables 2 to 4. In light and dark conditions, a significant difference in AOD500, TISA500, ARA, ACD, ACW, ACA, and ACV was found among the three groups (P < 0.05). Following pharmacologic dilation, there was a significant difference in AOD500, TISA500, ACD, ACW, ACA, and ACV among the three groups (P < 0.05). 

Table 2

 

AS-OCT Data of PACD Subjects With Different PAC Mechanisms in the Light

Parameter	1 = PB,n= 122	2 = PIC,n= 150	3 = TPIR,n= 144	PValue*	PValue, 1 & 2†	PValue, 1 & 3†	PValue, 2 & 3†
AOD500, mm (SD)	0.236 (0.099)	0.295 (0.099)	0.226 (0.104)	<0.001	<0.001	1.000	<0.001
TISA500, mm2 (SD)	0.102 (0.038)	0.120 (0.042)	0.093 (0.044)	<0.001	0.001	0.255	<0.001
ARA, mm2 (SD)	0.273 (0.105)	0.312 (0.115)	0.238 (0.116)	<0.001	0.017	0.037	<0.001
ACD, mm (SD)	2.116 (0.266)	2.385 (0.233)	2.234 (0.223)	<0.001	<0.001	<0.001	<0.001
ACW, mm (SD)	10.82 (0.44)	11.05 (0.41)	10.86 (0.39)	<0.001	<0.001	1.000	<0.001
ACA, mm2 (SD)	14.82 (2.66)	17.26 (2.26)	15.70 (2.29)	<0.001	<0.001	0.009	<0.001
ACV, mm3 (SD)	56.85 (13.17)	68.43 (11.43)	60.07 (12.05)	<0.001	<0.001	0.098	<0.001
PD, mm (SD)	3.86 (0.72)	3.95 (0.73)	4.05 (0.60)	0.085

*  The 1-way ANOVA.

†  Bonferroni tests.

Table 3

 

AS-OCT Data of PACD Subjects With Different PAC Mechanisms in the Dark

Parameter	1 = PB,n= 122	2 = PIC, n= 150	3 = TPIR,n= 144	PValue*	PValue, 1 & 2†	PValue, 1 & 3†	PValue, 2 & 3†
AOD500, mm (SD)	0.197 (0.100)	0.262 (0.102)	0.194 (0.112)	<0.001	<0.001	1.000	<0.001
TISA500, mm2 (SD)	0.085 (0.039)	0.104 (0.039)	0.078 (0.043)	<0.001	<0.001	0.427	<0.001
ARA, mm2 (SD)	0.215 (0.105)	0.260 (0.097)	0.196 (0.110)	<0.001	0.001	0.398	<0.001
ACD, mm (SD)	2.131 (0.292)	2.387 (0.225)	2.237 (0.225)	<0.001	<0.001	0.001	<0.001
ACW, mm (SD)	10.89 (0.44)	11.08 (0.35)	10.88 (0.40)	<0.001	<0.001	1.000	<0.001
ACA, mm2 (SD)	15.22 (2.83)	17.60 (2.27)	16.04 (2.35)	<0.001	<0.001	0.023	<0.001
ACV, mm3 (SD)	58.33 (14.04)	70.00 (11.55)	61.25 (12.23)	<0.001	<0.001	0.178	<0.001
PD, mm (SD)	4.56 (0.71)	4.65 (0.77)	4.77 (0.65)	0.053

*  The 1-way ANOVA.

†  Bonferroni test.

Table 4

 

AS-OCT Data of PACD Subjects With Different PAC Mechanisms Following Pharmacological Dilation

Parameter	1 = PB,n= 68	2 = PIC,n= 100	3 = TPIR,n= 96	PValue	PValue, 1 & 2	PValue, 1 & 3	PValue, 2 & 3
AOD500, mm (IR)	0.216 (0.176, 0.282)	0.281 (0.209, 0.359)	0.218 (0.158, 0.303)	<0.001†	<0.001§	0.939§	<0.001§
TISA500, mm2 (SD)	0.097 (0.034)	0.110 (0.040)	0.095 (0.041)	0.014*	0.085‡	1.000‡	0.020‡
ARA, mm2 (SD)	0.283 (0.110)	0.311 (0.122)	0.271 (0.124)	0.062*
ACD, mm (SD)	2.170 (0.263)	2.487 (0.212)	2.336 (0.232)	<0.001*	<0.001‡	<0.001‡	0.001‡
ACW, mm (SD)	11.03 (0.39)	11.26 (0.36)	11.08 (0.42)	<0.001*	0.001‡	1.000‡	0.004‡
ACA, mm2 (SD)	17.69 (2.62)	20.79 (2.23)	19.19 (2.42)	<0.001*	<0.001‡	<0.001‡	<0.001‡
ACV, mm3 (SD)	75.24 (14.13)	90.68 (12.51)	82.05 (13.51)	<0.001*	<0.001‡	0.004‡	<0.001‡
PD, mm (SD)	7.03 (0.49)	7.16 (0.71)	7.19 (0.53)	0.238*

*  The 1-way ANOVA.

†  Kruskal-Wallis test.

‡  Bonferroni test.

§  Mann-Whitney test (<0.05/3 = 0.017 = significantly different).

Iris Cross-Sectional Area Measurements

A summary of mean IAs from eyes of the three groups measured in light, dark, and after pharmacologic dilation is presented in Table 5. There was a significant difference in IA in light and dark (P ≤ 0.001) among the three groups. 

Table 5

 

IA of Subjects With Different PAC Mechanisms in Three Conditions

Parameter	1 = PB,n= 68	2 = PIC,n= 100	3 = TPIR,n= 96	PValue*	PValue, 1 & 2†	PValue, 1 & 3†	PValue, 2 & 3†
IA in the light, mm2 (SD)	2.81 (0.39)	2.86 (0.39)	3.02 (0.39)	<0.001	1.000	<0.001	0.003
IA in the dark, mm2 (SD)	2.67 (0.38)	2.67 (0.37)	2.84 (0.35)	0.001	1.000	0.001	0.001
IA following pharmacological dilation, mm2 (SD)	1.79 (0.30)	1.82 (0.35)	1.87 (0.30)	0.283*

*  The 1-way ANOVA.

†  Bonferroni test.

Table 6 summarizes changes in IA, IA loss per mm PD increase, PD, and the ratio of CCD increase relative to PD increase in the three groups between light to dark, light to pharmacologic dilation, and dark to pharmacologic dilation. Differences in the changes of IAs from light to pharmacologic dilation (P = 0.030), IA loss per mm PD increase from light to pharmacologic dilation (P = 0.001), and dark to pharmacologic dilation (P = 0.011) were observed among the three groups, and were statistically significant. Bonferroni corrected comparisons showed significant differences in the changes of IAs (P = 0.028) between the PB and TPIR groups. Compared to the PB and PIC groups, the TPIR group had the greatest IA loss per mm PD increase from light to pharmacologic dilation (P = 0.001 and 0.010) and from dark to pharmacologic dilation (P = 0.039 and 0.026). 

Table 6

 

Changes in IA and PD After Mydriasis in Subjects With Different PAC Mechanisms

Parameter	1 = PB	2 = PIC	3 = TPIR	PValue	PValue, 1 & 2	PValue, 1 & 3	PValue, 2 & 3
L to D, 278 eyes
IA change, mm2 (SD)	0.19 (0.17)	0.22 (0.18)	0.24 (0.24)	0.244*
IA loss per mm PD increase (IR)	0.231 (0.082, 0.351)	0.253 (0.127, 0.392)	0.282 (0.136, 0.385)	0.252†
PD change, mm (SD)	0.95 (0.35)	0.92 (0.32)	0.73 (0.47)	0.768*
Centroid increase/PD increase (SD)	0.434 (0.347)	0.398 (0.379)	0.363 (0.412)	0.452*
L to P, 261 eyes
IA change, mm2 (SD)	1.01 (0.31)	1.07 (0.34)	1.15 (0.33)	0.030*	0.829‡	0.028‡	0.278‡
IA loss per mm PD increase (IR)	0.313 (0.279, 0.368)	0.339 (0.300, 0.379)	0.362 (0.311, 0.430)	0.001†	0.082§	0.001§	0.010§
PD change, mm (SD)	3.17 (0.68)	3.20 (0.88)	3.14 (0.53)	0.840*
Centroid increase/PD increase (IR)	1.255 (0.878, 1.891)	1.207 (0.769, 1.646)	1.180 (0.859, 1.545)	0.392†
D to P, 264 eyes
IA change, mm2 (SD)	0.87 (0.27)	0.87 (0.30)	0.96 (0.30)	0.059*
IA loss per mm PD increase (SD)	0.361 (0.114)	0.363 (0.105)	0.408 (0.129)	0.011*	1.000‡	0.039‡	0.026‡
PD change, mm (SD)	2.46 (0.67)	2.43 (0.83)	2.41 (0.50)	0.908*
Centroid increase/PD increase (IR)	1.004 (0.506, 1.352)	0.882 (0.356, 1.213)	0.880 (0.449, 1.236)	0.380†

 L to D, light to dark; L to P, light to after pharmacological dilation; D to P, dark to after pharmacological dilation.

*  The 1-way ANOVA.

†  Kruskal-Wallis test.

‡  Bonferroni test.

§  Mann-Whitney test (<0.05/3 = 0.017 = significant different).

In eyes with PB, partial correlation coefficients were −0.800 for IA and PD. Partial correlation coefficients were −0.842 for IA and PD in eyes with PIC. In eyes with TPIR, Pearson correlation coefficient for IA and PD was −0.842. Linear regression analysis showed that IA decreased with increasing PD over the whole range (light to dark and pharmacologic dilation) in all three groups (Fig. 5). Change in the IA with increase in PD (slope = 0.315 for PB, 0.324 for PIC, and 0.353 for TPIR) were significantly different between PB and TPIR groups (P = 0.040), but not significantly different between PB and PIC (P = 0.613), PIC, and TPIR (P = 0.076) groups. 

Figure 5

 

Pupillary diameter versus IA. PB y = −0.315x + 4.061 (R² = 0.647, P < 0.001, linear regression). PIC y = −0.324x + 4.178 (R² = 0.705, P < 0.001, linear regression). TPIR y = −0.353x + 4.460 (R² = 0.708, P < 0.001, linear regression). F = 0.506, P = 0.613 (pupillary block versus plateau iris configuration). F = 2.063, P = 0.040 (pupillary block versus thick peripheral iris roll). F = 1.782, P = 0.076 (plateau iris configuration versus thick peripheral iris roll).

Results of univariable and multivariable linear regression analysis of IA change from light to dark are shown in Table 7. Younger age (P < 0.001), larger PD changes (P < 0.001), and lager CCD distance in the light (P = 0.011) were associated with greater decrease in IA. 

Table 7

 

Factors Associated With IA Change (Light to Dark)

Variable	Univariable		Multivariable
	β	PValue	Direction	β	PValue
Age	−0.177	0.003	Younger subjects had more IA decrease	−0.203	<0.001
Sex	−0.015	0.800		–	–
SE	−0.133	0.055		–	–
CCT	−0.014	0.823		–	–
ACD	0.087	0.155		–	–
LT	−0.085	0.164		–	–
AL	−0.059	0.336		–	–
AC mechanism: PB vs. PIC	0.010	0.862		–	–
AC mechanism: PB vs. TPIR	0.079	0.190		–	–
PD; L	−0.189	0.002		−0.0032	0.971
PD; D	0.059	0.323		–	–
PD change; L to D	0.474	<0.001	Eyes with larger PD change (L to D) had more IA decrease	0.452	<0.001
CCD; L	−0.158	0.008	Eyes with larger CCD (L) had more IA decrease	−0.137	0.011
CCD; D	0.115	0.055		–	–
CCD increase; L to D	0.330	<0.001		0.072	0.264
CCD increase/PD increase; L to D	0.152	0.011		0.048	0.404

 L, light; D, dark.

Discussion

The normal iris may freely gain or lose extracellular water from the spongy stroma and loses volume after physiological or pharmacological pupil dilation by this mechanism.³⁹ The iris of PACS and PACD may have more compact or water-retentive stroma (poor fluid conductivity) that would lose less fluid and thus retain more volume with pupil dilation.^9,10,18,40 

Pure pupillary block, pure nonpupillary block, and combination of multiple mechanisms have been reported to underlie AC in Chinese eyes with PACD and fit in with the multifactorial nature of disease.²³ Our study is an initial attempt to obtain insights into the relationship between different AC mechanisms and dynamic iris changes in the pathogenesis of PACS and PACD. Accordingly, while the eyes in our study had more than one mechanism underlying AC we identified what we felt was the dominant mechanism (PB, PIC, and TPIR) present in each eye. We found differences in iris behavior following dilation (both physiologic and pharmacologic) in eyes with these different AC mechanisms. 

Compared to pharmacologic dilation, IA loss per mm PD increase decreased in dark and light in all three groups, with the smallest decrease in the PB group. While the PD change was similar, the change in IA per mm change in PD was significantly different between all three groups for the comparison between light and pharmacologic dilatation, the change in condition that produced the largest change in pupil size. The trend for the PB group loosing the least IA area can be seen with the change in pupil size in different conditions: from light to dark, dark to pharmacological dilatation, and finally with light to pharmacological dilation. The IA decreased relatively linearly with increasing PD in all eyes: the PB group had the smallest slope and the TPIR group the largest. The results suggested that dynamic iris change with pupil dilation may have a more important role in the pathogenesis of PACD in eyes with PB as the dominant mechanism for AC, compared to PIC or TPIR. 

In PACS and PACD eyes with PB as the dominant AC mechanism, there is resistance to aqueous flow from the posterior to anterior chamber at the level of the pupil. This results from three forces, sphincter and dilator muscles, as well as iris elasticity, creating a pressure gradient that causes bombe of the peripheral iris and closure of the angle.^41,42 The pressure gradient between anterior and posterior chamber may affect the iris structure and change the capacity for free fluid movement. Almost all eyes with PACD are likely have some degree of pupillary block.⁴³ The contribution of PB, while possibly a universally necessary cause, may be less or require a different set of complementary causative factors where PIC or TPIR is the dominant mechanism. A lower pressure gradient between the anterior and posterior chambers with less effect on capacity for transfer of fluid may explain some of the differences in dynamic iris change among the three groups. 

Previous studies have reported PB as the main mechanism responsible for AAC.^26,27 There is no significant difference in ACD, LT, and AL between the AAC and contralateral eye⁴⁴; in such fellow eyes of Caucasian patients IV increased following physiologic or pharmacologic pupil dilation.⁹ However, the reported IV increase was due to an error in the calculation formula where greater CCD change with a decreased IA may result in an increased IV. 

Interestingly, subjects with PIC had a better PVA than those with PB. This was due to the latter having a relatively higher hypermetropic refractive error and is another indication of the multiple components and the complexity of AC. 

The PACD is clearly a multifactorial disease. The main weakness in our study is the assumption that there may be a dominant mechanism involved and the method of determination of this dominant mechanism. The latter was done using forced choice by a trained observer and clearly there is a subjective element involved. It could be argued that the findings are a mere reflection of the forced choice. In the absence of more objective methods, we feel the use of a forced choice by an experienced observer is acceptable as an initial attempt to determine potential differences (if any) in dynamic iris changes between the dominant mechanisms hypothesized. Such a selection does have prior plausibility and is supported by the sufficient component model of causation used to identify the role of individual mechanisms in causation in a multifactorial disease like PACD.^6,7,43 The use of the AS-OCT alone to determine the mechanism is another weakness, but our earlier unpublished study found a good k with the UBM for the same observer. In the absence of better methods to determine the dominant mechanism the data from this initial study must be considered preliminary while methods to quantify the identification of the dominant mechanism, if any, are developed. The way forward should include longitudinal studies to determine if the dynamic IA loss per mm PD increase is a significant predictor for progression of PACS (increasing IOP, formation of PAS, or AAC) that can help identify patients who need treatment. The IA loss per mm PD increase as a possible predictive test will be more practical if the safer approach of comparing bright to dim illumination is used, since there will be a natural reluctance to pharmacologically dilate high-risk suspects. Continued planned follow-up of patients in this study may provide such information, but larger numbers likely will be needed. 

In conclusion, we have presented preliminary insights into the dynamic iris change in PACS and PACD eyes with different dominant AC mechanisms in a rural Chinese population. The data suggested that iris dynamic change may have a more important role in PACD eyes with PB, compared to those with PIC or TPIR. 

Acknowledgments

The authors thank Jingwei Zheng for his assistance with the statistical analysis, all staff who contributed to this study, and all the subjects who participated. 

Supported by a research special fund of Ministry of Health of the people's republic of China, Grant Number 201002019. The authors alone are responsible for the content and writing of the paper. 

Disclosure: Y. Zhang, None; S.Z. Li, None; L. Li, None; M.G. He, None; R. Thomas, None; N.L. Wang, None